Returns whether there is an open or a closed eye given an image from surrounding area with about 99% accuracy.

See https://www.kaggle.com/code/dima806/closed-eye-image-detection-vit for more details.

Classification report:

              precision    recall  f1-score   support

    closeEye     0.9921    0.9888    0.9904      4296
     openEye     0.9889    0.9921    0.9905      4295

    accuracy                         0.9905      8591
   macro avg     0.9905    0.9905    0.9905      8591
weighted avg     0.9905    0.9905    0.9905      8591

Downloads last month: 561

Safetensors

Model size

85.8M params

Tensor type

F32

Model tree for dima806/closed_eyes_image_detection

Base model

google/vit-base-patch16-224-in21k

Finetuned

(2526)

this model

dima806
/

closed_eyes_image_detection

Model tree for dima806/closed_eyes_image_detection

Spaces using dima806/closed_eyes_image_detection 5